riders on the storm - home, the social market … · riders on the storm contents acknowledgements...
TRANSCRIPT
Nida Broughton, Onyinye Ezeyi, Claudia Hupkau
Britain’s middle income households since 2007
RIDERS ON THE STORM
THE SO
CIA
L MA
RK
ET FOU
ND
ATIO
NR
IDER
S ON
THE STO
RM
: BR
ITAIN
’S MID
DLE IN
CO
ME H
OU
SEHO
LDS SIN
CE 2007
N
IDA
BR
OU
GH
TON
, ON
YIN
YE EZ
EYI, C
LAU
DIA
HU
PKA
U
Copyright © Social Market Foundation, 2014
ISBN: 978-1-904899-89-1
£10.00
Nida Broughton, Onyinye Ezeyi and Claudia Hupkau
Britain’s middle income households since 2007
RIDERS ON THE STORM
FIRST PUBLISHED BY The Social Market Foundation, April 2014 ISBN: 978-1-904899-89-1
11 Tufton Street, London SW1P 3QBCopyright © The Social Market Foundation, 2014The moral right of the authors has been asserted. All rights reserved. Without limiting the rights under copyright reserved above, no part of this publication may be reproduced, stored or introduced into a retrieval system, or transmitted, in any form or by any means (electronic, mechanical, photocopying, recording, or otherwise), without the prior written permission of both the copyright owner and the publisher of this book.
THE SOCIAL MARKET FOUNDATIONThe Foundation’s main activity is to commission and publish original papers by independent academic and other experts on key topics in the economic and social fields, with a view to stimulating public discussion on the performance of markets and the social framework within which they operate.
The Foundation is a registered charity and a company limited by guarantee. It is independent of any political party or group and is funded predominantly through sponsorship of research and public policy debates. The Foundation takes complete responsibility for the views expressed in this publication, and these do not necessarily reflect the views of the sponsor.
CHAIRMANMary Ann Sieghart
MEMBERS OF THE BOARDDaniel FranklinLord John HuttonGraham MatherSir Brian Pomeroy cbeBaroness Gillian ShephardNicola HorlickProfessor Alison Wolf
DIRECTOREmran Mian
DESIGN AND PRODUCTIONSoapbox www.soapbox.co.uk
PRINTED BY Repropointwww.repropoint.com
RIDERS ON THE STORM
CONTENTS
Acknowledgements 4
About the authors 5
Executive summary 6
1. WHY WORRY ABOUT MIDDLE INCOME HOUSEHOLDS? 10
2. WHO IS IN THE MIDDLE, AND HOW HAS THE MIDDLE CHANGED? 13What the middle looks like 13
How the middle has changed since 2007 15
3. COPING WITH THE DOWNTURN 20Keeping money flowing in 20
Keeping costs under control 21
The future squeeze 26
“Are you better off than you were four years ago?” 28
4. HELPING MIDDLE INCOME HOUSEHOLDS 30Comparing and switching 32
Giving people access to the right information 33
Liberating data 34
Just pushing data is not enough 36
ANNEX: METHODOLOGY 39Overview 39
Analysis of Understanding Society and the
British Household Panel Study 40
Analysis of the Family Resources Survey 42
Analysis of the Living Costs and Food Survey 43
SOCIAL MARKET FOUNDATION
4
ACKNOWLEDGEMENTS
The publication of this report has been made possible by the
generous support of Santander.
Our Families Panel, established to advise on the SMF’s
programme of work on middle income families, provided valuable
thoughts and insights on our initial data findings. Thanks are due
to Jonathan Akerman, Chris Bose, Elizabeth Duff, Duncan Hames,
Adrian Harvey, Matt Oakley and Sue Willis. We would also like to
thank Which? for providing data extracts from their Consumer
Insight Tracker.
At the SMF, we would like to thank Emran Mian and Nigel
Keohane for intellectual input throughout the project.
RIDERS ON THE STORM
5
ABOUT THE AUTHORS
NIDA BROUGHTON
Nida Broughton is Chief Economist at the SMF, where she has led
research on skills policy, entrepreneurship and analysis of public
spending, as well as undertaking research on a range of public policy
areas including healthcare, education and housing. Nida previously
worked at the House of Commons, where she advised MPs and
committees on a broad range of economic issues, and in particular,
on financial services. Nida also has in-depth expertise in consumer
regulation, competition policy and behavioural economics, gained
whilst working at Ofcom, the UK regulator and competition authority
for communications markets. She has an MA (Cantab) in Economics
from Cambridge University and an MSc in Economics from Birkbeck
College, University of London.
ONYINYE EZEYI
Onyinye Ezeyi is a Research Intern at the Social Market Foundation.
She recently completed an MSc in Development Economics from
the University of Sussex after graduating with a First in BSc (Hons)
Economics with Sociology from Loughborough University in 2011.
Onyinye’s interests lie in the use of econometric methodologies in
social research and has worked on projects in migration and health.
She plans to study for a PhD in Economics.
CLAUDIA HUPKAU
Claudia Hupkau is a Research Fellow for the Social Market
Foundation. She is currently completing a PhD in Economics at the
Université Catholique de Louvain (UCL), Belgium, and is a member
of the Centre for Operations Research and Economics, Belgium. She
is part of the research group ‘Poverty, Resource Equality and Social
Policy’ at UCL and has worked on the behavioural economics of
poverty and education.
SOCIAL MARKET FOUNDATION
6
EXECUTIVE SUMMARY
In the years since the financial crisis, households across the income
distribution have seen a combination of wage stagnation and a rise
in the cost of living. Working age households in particular have been
adversely affected by real declines in their main source of income.
Much analysis has focused on lower income households. But
more recently, the so-called “cost of living crisis” has pushed the
political focus onto the “squeezed middle”. Households in the
middle and fourth quintile of the income distribution have annual
incomes between around £26,100 and £63,000. They receive some
cash benefits, but given the pressure to reduce public spending,
they are unlikely to be prime candidates for more state transfers,
leaving them more reliant on their own resources in coping with
the recent economic climate.
So how well have they fared? Is it indeed the case that they
have been squeezed? How successfully have they coped with
rising prices whilst incomes have stagnated?
The SMF’s analysis draws on the British Household Panel and
Understanding Society surveys to follow specific households through
time, from the brink of the downturn in 2007–08 to 2011–12, the latest
wave of available data. We complement this with data from other
national surveys to look at specific types of spending in more detail.
What is most striking is that middle income households today
are not the same households that were in the middle going into
the downturn. The next most striking finding is the very diverse
experiences of the middle. Of those who were in the middle in
2007–08:
• 6% had dropped into the bottom fifth of the distribution by
2011–12
RIDERS ON THE STORM
7
• Just over 10% had climbed up to the top fifth of the
distribution by 2011–12
• Overall the households that climbed up the distribution
outnumbered those that fell.
Correspondingly, today’s middle income households have
seen a diverse range of journeys to where they are now. Of those
households in the middle quintile in 2011–12:
• One in 10 were in the bottom fifth of the income distribution
four years earlier
• One in five were in the second to bottom quintile four years
earlier
• 17% were previously higher up the distribution and dropped down.
Overall, this combination of climbers and fallers meant that
those in the middle by 2011–12 saw their average real income
remain roughly the same since 2007–08. Meanwhile, the quintile
just above – the fourth quintile – saw a real increase. In contrast,
those who were in the bottom two quintiles by 2011–12 had seen
falls in real income.
Moving up into and remaining in the middle has been made
possible through higher employment levels. Households in the
middle today are more likely to have two earners rather than one
compared to 2007–08. They have managed spending on food –
switching to cheaper items – to avoid the 25% increase in food
prices over the period. Childcare is still problematic for many, with
a heavy reliance on grandparents compared to those on higher
incomes. But they have also been helped by low interest rates,
which have kept housing costs down among the large number of
home owners in this category.
For many middle income households, there are rising costs to
come when interest rates go up again. With competing pressures
SOCIAL MARKET FOUNDATION
8
on public spending, there is less scope for more state help. But
this does not mean that nothing can be done. One area of policy
where these households can be helped to do more with less is
in consumer markets. Those in the middle today have displayed
significant resilience thus far in coping with the downturn. But
there are markets where more needs to be done to help these
households help themselves in securing better value for money.
Substantial progress has already been made on making
switching processes easier. Now, the bigger gains are likely to come
from helping consumers easily compare products and services.
This means ensuring there is sufficient information easily available,
that product features and pricing are transparent, and that it is easy
to use information that does exist to calculate likely savings from
switching.
For middle income families in particular, managing childcare
costs is difficult. This is not helped by the fact that childcare is one
area where information that allows parents to compare providers is
very sparse. Whilst there is some information on quality, comparing
prices and value for money is much harder. Helping parents access
this information will help them make better choices about what
childcare is best for them.
There is also a need to ensure that consumers have access to
the right information and that product features and pricing are
transparent. As interest rates rise again, the middle is likely to see
further pressure on their spending. Last year saw rises in tracker
mortgage rates despite the Bank of England base rate remaining
the same. It is vital that there is sufficient clarity from banks and
regulators as to what consumers should expect from different
types of products so that they can make informed choices about
how to manage the future rise in mortgage costs. Measures such as
the implementation of the Financial Conduct Authority’s Mortgage
Market Review will help with this.
RIDERS ON THE STORM
9
Finally, making information easy to access and use to calculate
savings is vital. Consumers could greatly benefit if they had better
access to their own data to compare prices and providers, as is
happening through the midata initiative. Even better would be the
development of services that allow consumers to compare how much
they are spending compared to the average, as an indicator of the
level of savings possible to achieve by switching. Other products and
services – beyond energy, mobile phones, current accounts and credit
cards – should be considered as part of the midata strategy. These
include telecoms and broadband more generally; savings products
and groceries. Government, regulators and industry will need to
ensure that major players in the industry act together, whether this is
achieved through voluntary agreement or compulsion.
But pushing data alone will not always be enough. Even when
the right information is available and easily comparable, evidence
suggests that prompts may be needed to encouraging switching.
New, innovative apps that scan the market and alert consumers to
new deals may be developed. Where this does not happen, other
measures should be considered. The model of renewal notices
in the car insurance market, which consumers say prompts them
to shop around, could be considered for other sectors, and if
combined with easy to use usage data, could be very powerful.
Given that Government is less than halfway through its
spending cuts programme, there will be many competing
pressures on public expenditure. But this does not mean that
middle income households cannot be helped. These households
have shown remarkable resilience so far, demonstrating that
under the right circumstances, and with the right information to
hand they can manage their finances and cope with being under
pressure. There should be greater focus on equipping the middle
to negotiate consumer markets where it is harder to secure better
value for money. Making this happen is likely to require action from
Government, regulators and industry.
SOCIAL MARKET FOUNDATION
10
1. WHY WORRY ABOUT MIDDLE INCOME HOUSEHOLDS?
Households across the income distribution have seen a combination
of wage stagnation and rises in the cost of living since 2008. The
period 2010 to 2013 in particular is the longest sustained period of
falling real wages in the UK on record.1
Working age households have borne the brunt, and specifically
have been adversely affected by the fall in real wages. Among these
working age households, those in the middle fifth receive less state
help in the form of benefits compared to lower quintiles, although
they do receive some, as shown in Chart 1.1. Instead, they have been
more reliant on their own resources to cope with the downturn.
Chart 1.1: Benefits and taxes as a proportion of original income
0%
20%
-20%
40%
-40%
60%
-60%
80%
-80%
100%
Cash benefits
Quintile 1 Quintile 2 Quintile 3 Quintile 4 Quintile 5
Direct taxes Indirect taxes
Office for National Statistics, “The effect of taxes and benefits on household income, 2011–12”, released 10 July 2013
Note: Non-retired households, quintile groups ranked on equalised disposable income
The middle, and to some extent the fourth, quintile is of specific
interest. Unlike the very top quintile, they are less likely to have
been able to accumulate savings to help negotiate the downturn.
1 Office for National Statistics, “An examination of falling real wages, 2010–2013”, 31 January 2014, http://www.
ons.gov.uk/ons/dcp171766_351467.pdf
RIDERS ON THE STORM
11
The middle is also subject to broader economic and long-term
pressures driven by globalisation and technological change. There
has been a “hollowing out” effect, whereby the proportion of jobs in
low wage and high wage occupations has risen and the proportion
in middle wage occupations has fallen. Whilst this effect is likely to
be relatively small on a year-to-year basis, evidence suggests that
the recession accelerated the decline in middle wage jobs.2
This report examines the reality of the “squeezed middle”:
looking at who the middle are, where they were in 2007–08 and
how they coped with the downturn. To do this, we use longitudinal
data – from the British Household Panel (BHPS) and Understanding
Society (US) surveys,3 that follow specific households over time.
We compare income and spending in two years: 2007–08, on the
brink of the downturn, and 2011–12, the latest data available. Where
longitudinal data is unavailable, we complement this analysis with
data from the Family Resources Survey (FRS)4 and the Living Costs
and Food Survey (LCFS).5 Full detail on methodology and sources
are included in the Annex to this report.
We find that those in the middle by 2011–12 were precisely
those families that had weathered the downturn relatively well.
Employment for these households held up – and sometimes
increased – and they were able to manage spending to minimise
the effect of rising prices. But many of these households will see
rising costs in the future; when interest rates rise, mortgage costs,
which have been kept low, are likely to rise again. The final section
2 Paul Sissons, The hourglass and the escalator (London: The Work Foundation, 2011)
3 University of Essex. Institute for Social and Economic Research and National Centre for Social Research,
Understanding Society: Wave 3, 2011–2012 Colchester, Essex: UK Data Archive SN: 6614; British Household Panel
Survey: Wave 17, 2007–08 Colchester, Essex: UK Data Archive SN: 5151.
4 Department for Work and Pensions, National Centre for Social Research and Office for National Statistics.
Social and Vital Statistics Division, Family Resources Survey, 2007–08 and 2010–2011. Colchester, Essex: UK Data
Archive, SN: 7085 , http://dx.doi.org/10.5255/UKDA-SN-7085-1
5 Office for National Statistics and Department for Environment, Food and Rural Affairs, Living Costs and Food Survey,
2011 Colchester, Essex: UK Data Archive, August 2013. SN: 7272 , http://dx.doi.org/10.5255/UKDA-SN-7272-2
SOCIAL MARKET FOUNDATION
12
of this report looks at what policymakers can do to ease financial
pressures in the future.
The report is set out as follows:
• Chapter 2 sets out what the middle looks like, and how it has
changed since 2007–08, looking at trends in income.
• Chapter 3 shows how middle income households coped with
the downturn, by keeping money flowing in and managing
spending. It also looks at the effect of the downturn on their
well-being and prospects for the future.
• Chapter 4 discusses the main policy lever available to help
middle income households given the tightening in public
finances: making consumer markets work better.
RIDERS ON THE STORM
13
2. WHO IS IN THE MIDDLE, AND HOW HAS THE MIDDLE CHANGED?
This chapter sets the scene, showing what the middle looks like in
terms of employment and occupations, and how the middle has
changed since the downturn.
In our analysis, we focus on working age middle income
households, defined as the middle fifth or middle quintile. We also
look at the quintile just above, quintile 4. More specifically, we rank
households by their original income, before taxes and transfers,
from the top income households in quintile 5 to the bottom
income households in quintile 1. The purpose of using the original
income measure as opposed to other measures of income is to
better understand the underlying trends and changes that middle
income families underwent.
In our dataset in 2011–12, middle quintile households had
an average annual original income ranging between £26,100
and £41,200. On average, original income of these households
was £33,600. Those in the fourth quintile had an average annual
income of £51,100, with a range of between £41,200 and £63,000.
This section sets out the key characteristics of middle income
households, based on analysis of the BHPS and US.
WHAT THE MIDDLE LOOKS LIKE
The majority of middle income households – 91% – had at least
one person in employment in 2011–12. In addition, around half
of these households had two wage earners. Looking at the
proportion of doubler-earner households across the income
distribution in Chart 2.1, it is clear that having dual income sources
from two earners is a key factor that helps households reach the
higher end of the distribution. Where households in the middle
fifth differ from the quintile just below them is in the much higher
SOCIAL MARKET FOUNDATION
14
rate of double-earner households: 50% compared to just 21% in
the second quintile.
In contrast, the fourth quintile has even higher rates of double-
earner households compared to the middle – 64% . This is likely
to be at least partly due to the fact that whilst middle and fourth
quintile households are as likely as each other to have children,
the middle quintile are slightly more likely to have children under
10. The presence of younger children may make it harder for
households in the middle to have two adults working, and as we
will see later on, is likely to create particular childcare challenges for
these households.
Chart 2.1: Proportion of households with individuals in employment
100%90%80%70%60%50%40%30%20%10%0%
Quintile 5
Quintile 4
Quintile 3
Quintile 2
Quintile 1
Proportion of households with two earnersProportion of households with at least one person employed
SMF analysis of Understanding Society, 2011–12, Wave 3
In terms of occupations, the middle quintile has much in
common with the quintiles below. A substantial number of
individuals in the middle and fourth quintiles are in occupations
likely to be low-wage. Half of individuals in middle income
households are in the four lowest skill occupations, as shown in
Chart 2.2. As such, those in the middle are strongly affected by
labour market conditions in the lower wage occupations.
RIDERS ON THE STORM
15
Chart 2.2: Proportion of individuals split by occupation
Quintile 5
Quintile 4
Quintile 3
Quintile 2
Quintile 1
Managers, directors and senior officialsProfessional occupationsAssociate professional and technical occupationsAdministrative and secretarial occupationsSkilled trades occupations
Caring, leisure and other service occupationsSales and customer service occupationsProcess, plant and machine operativesElementary occupations
100%80%60%40%20%0%
SMF analysis of Understanding Society, 2011–12, Wave 3
HOW THE MIDDLE HAS CHANGED SINCE 2007
Real wages have seen a substantial decline following the crisis and
downturn. But in the last few years, different households have had
varying experiences of the downturn. This means that in addition to the
overall stagnation of wages, the shape of British society has changed.
Chart 2.3 shows the UK’s income distribution for working age
households in 2007–08 and 2011–12. While at the very top the
distribution has hardly changed, the middle and bottom have
shifted to the left, meaning that real incomes have decreased
for all but the top deciles. As of 2011–12, the average household
in the middle quintile had real income of 13% per cent less than
the middle quintile in 2007–08 – before taxes and benefits, as
real earnings fell.6 Because incomes have become more skewed
6 The ONS has calculated that disposable income of non-retired middle quintile fell by 6.4%. Disposable
income is the amount of money that households have available for spending and saving after direct taxes
(such as income tax and council tax) have been accounted for. It includes earnings from employment, private
pensions and investments as well as cash benefits provided by the state; Office for National Statistics, “Middle
Income Households, 1977–2011/12”, 2 December 2013, http://www.ons.gov.uk/ons/dcp171776_341133.pdf
SOCIAL MARKET FOUNDATION
16
towards the bottom, it is now “easier” to get into the middle fifth in
terms of absolute income required.
Chart 2.3: UK income distribution in 2007–08 and 2011–12
0
.02
.04
.06
.08
Frac
tion
0 5,000 10,000 15,000
0
.02
.04
.06
.08
Frac
tion
0 5,000 10,000 15,000
Monthly household income before benefits
Monthly household income before benefits
2007–08
2011–12
SMF analysis of Family Resources Survey 2007–08 and 2011–12. Restricted to incomes between 0–15000 to
exclude outliers. 2011–12 prices
This has a number of implications for attempting to understand
how the middle has changed. The fall of many people to the
bottom of the distribution means that households in 2011–12
needed a lower income to be classified as in the middle. The gap
between low and middle income has narrowed.
If all household incomes had shifted downward by the same
amount, there would be no difference between the households
that were in the middle in 2007–08 and those that were in the
middle in 2011–12. But this was not the case. Many of the households
RIDERS ON THE STORM
17
that previously formed part of the middle have themselves moved
elsewhere in the distribution, as shown in Chart 2.4. The middle of
2007–08 is not the same as the middle of 2011–12.
Chart 2.4: Where the 2007–08 middle and fourth quintiles went
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
Quintile 1, 2011–12Quintile 4, 2011–12
Quintile 2, 2011–12Quintile 5, 2011–12
Quintile 3, 2011–12
Quintile 3, 2007–08 Quintile 4, 2007–08
SMF analysis of British Household Panel Survey, 2007–08, Wave 17; Understanding Society, 2011–12, Wave 3
In fact, one in 10 of those classified as being in the middle in
2007–08 made it all the way to the top quintile four years later,
and 31% made it into the fourth quintile. 6% fell into the bottom
quintile, but overall, the “climbers” from the middle income quintile
outnumbered the “fallers”.
What of the households that were in the middle coming out
of the downturn? The so-called “squeezed middle” of 2011–12 is
a mixed group. Over a quarter were made up of households that
used to be in the top two quintiles. But there has been remarkable
upward movement too, with one in 10 of them coming from the
bottom quintile, as shown in Chart 2.5.
SOCIAL MARKET FOUNDATION
18
Chart 2.5: Where the 2011–12 middle and fourth quintiles came from
Quintile 3, 2011–12
Quintile 4, 2011–12
100%90%80%70%60%50%40%30%20%10%0%
Quintile 1, 2007–08Quintile 4, 2007–08
Quintile 2, 2007–08 Quintile 3, 2007–08Quintile 5, 2007–08
SMF analysis of British Household Panel Survey, 2007–8, Wave 17; Understanding Society, 2011–12, Wave 3
Overall, this combination of climbers and fallers means that
those in the middle as of 2011–12 saw – on average – no significant
change in income. It has become in a sense easier to enter the
middle quintile: the income threshold is lower than it used to be.
This means that although the climbers outnumber the fallers, the
climbers did not require as big an income rise as they would have
previously needed to be considered middle income. But as we shall
see in the next chapter, employment played a significant part in
helping households climb into and remain in the middle.
At first this may seem counter-intuitive given the fall in real
income between the middle quintile in 2007–08 compared to the
middle quintile in 2011–12. But this real income fall is a snapshot
analysis, that does not reflect the fact that many of those in the
middle by 2011–12 were not in the middle four years previously: they
were the downturn’s success stories, moving up to the middle from
the bottom quintiles. Tracking actual households over time allows
us to better understand where the middle of today comes from,
compared to a snapshot analysis. The households in the middle by
2011–12 were often the households that had seen improvements in
their financial situation compared to the average.
RIDERS ON THE STORM
19
Chart 2.6: Original monthly income before taxes and benefits
(2011–12 prices)
£5,000
£4,500
£4,000
£3,500
£3,000
£2,500
£2,000
£1,500
£500
£0
£1,000
2007–08 2011–12
Quintile 3 in 2011–12 Quintile 4 in 2011–12
£2,961
£3,909
£2,817
£4,253
SMF analysis of British Household Panel Survey, 2007–08, Wave 17; Understanding Society, 2011–12, Wave 3
Error bars represent 95% confidence intervals
Quintiles are defined based on the 2011–12 income distribution. The chart compares the incomes of the same
group of households in 2007–08 and 2011–12.
The rest of this report looks at how those in the middle today
have coped with the downturn – how they have managed to
maintain their incomes and keep a lid on spending during a time
of rising prices.
SOCIAL MARKET FOUNDATION
20
3. COPING WITH THE DOWNTURN
This chapter looks at how those in the middle today coped with the
downturn, and how this affected their well-being.
KEEPING MONEY FLOWING IN
Those who were younger when the recession hit were more likely
to have moved up to the middle quintile by 2011–12 – most likely
due to the fact that they were on low wages at the start of their
careers, and saw income increases as they progressed on. But this
“career lifecycle” effect explains the smaller part of the change in
households’ relative positions.
Employment – and especially employment of a second
household member – is the main driver affecting a household’s
position in the income distribution. Many middle income
households in 2007–08 which lost a second earner fell down the
income scale. In the group of households that climbed up from the
bottom two quintiles into the middle, the proportion of double-
earner households rose from 35% to 49%. In contrast the group
that fell down from the top quintiles into the middle saw their
proportion of double earner households drop from 72% to 62%.
Overall, one in 20 of those in the middle by 2011–12 had
become a double-earner household over the preceding four years.
The majority of these new second earners – two thirds – had full
time jobs.
RIDERS ON THE STORM
21
Chart 3.1: Proportion of households with two earners
70%60%50%40%30%20%10%0%
Quintile 1in 2011–12
Quintile 2in 2011–12
Quintile 3in 2011–12
Quintile 4in 2011–12
2007–08
2011–12
2007–08
2011–12
2007–08
2011–12
2007–08
2011–12
2007–08
2011–12
Quintile 5in 2011–12
SMF analysis of British Household Panel Survey, 2007–08, Wave 17; Understanding Society, 2011–12, Wave 3
KEEPING COSTS UNDER CONTROL
EssentialsMiddle income households spend around 30% of their gross income
on essentials – food, mortgage and rent, and energy. Mortgage and
rent is the largest of these, and varies substantially across the UK.
Those on middle incomes in the North East and North West spend
about 15% of their gross income on these housing costs; in London,
this rises to 22%.
Chart 3.2: Spending on essentials as a proportion of gross income, 2011–12
Quintile 5
Quintile 4
Quintile 3
Quintile 2
Quintile 1
Food Energy Mortgage/rent
60%40%20%0% 10% 30% 50%
SMF analysis of Understanding Society, 2011–12, Wave 3
SOCIAL MARKET FOUNDATION
22
But during the downturn, middle income households have been
protected from rising housing costs. Over half of these households
own a house with a mortgage, and 15% own outright. Historically
low interest rates mean that their housing costs have either remained
steady or declined since 2007. The Bank of England base rate was over
5% in 2007; in contrast it has been 0.5% since 2009.
Where households have felt the pinch is on food. Food
prices rose by 25% from 2007 to 2011 and the cost of meals from
restaurants and cafes rose by 16%.7 However, middle income
households cut back on expensive items and switched to cheaper
ones, keeping the spending rise across groceries and meals to 10%
– below inflation. Households that fell down from the top quintiles
kept the food spending rise even lower.
They have also felt the pinch on energy. Whilst we cannot
compare energy spend directly because of temperature differences
in the two years we look at, we do know that the price of electricity,
gas and fuel rose by 35% from 2007 to 2011, on the CPI measure
of inflation. Data from the ONS (not specific to middle income
households) suggests that spending increased sharply over the
period 2007 to 2009. After 2010 total spending remained steady
– but partly due to reduced consumption as a result of milder
winters. More generally, over the past decade, households have
been reducing energy consumption through energy efficiency
measures and simply using less in response to higher bills.8
A similar story is apparent for transport spending, where costs
have increased by around 24% over the period we look at, on the CPI
measure of inflation. ONS analysis of general household spending
7 ONS, “Consumer Price Inflation Reference Tables”, January 2014, http://www.ons.gov.uk/ons/rel/cpi/
consumer-price-indices/january-2014/consumer-price-inflation-reference-tables.xls.
8 Office for National Statistics, “Household energy spending in the UK, 2002–2012”, 3 March 2014, http://
www.ons.gov.uk/ons/rel/household-income/expenditure-on-household-fuels/2002---2012/full-report--
household-energy-spending-in-the-uk--2002--2012.html
RIDERS ON THE STORM
23
finds that spending on transport has seen a general decline over
the past decade. This is likely to be at least partly due to consumers
responding to higher fuel prices by cutting car use and switching
to public transport; and switching to more energy efficient cars.9
In many middle quintile households, two adults go out to
work. But with less income than the top quintile, middle income
households are the most reliant on informal sources of childcare such
as grandparents, as shown in Chart 3.3. Those on higher incomes, in
the top quintile, are in contrast able to rely more on formal childcare.10
Chart 3.3: Proportion of households using grandparents for childcare
60%
50%
40%
30%
20%
10%
0%
2007–08 2011–12
Quintile 1 Quintile 2 Quintile 3 Quintile 4 Quintile 5
SMF analysis of Family Resources Survey 2007–08 and 2011–12
Whilst the proportion that rely on grandparents has slightly fallen
since 2007, those who rely on informal care have been using more
of it, at least in part due to rising childcare costs. The effect is most
clear for those with children under the age of two. Charts 3.4 and
3.5 show trends in spending and hours used. Whilst parents in the
middle quintile have barely increased the number of paid-for hours
9 Office for National Statistics, “Family Spending”, 11 December 2013, http://www.ons.gov.uk/ons/rel/family-
spending/family-spending/2013-edition/index.html
10 Note for childcare costs, we rely on FRS data. FRS does not follow households longitudinally. Comparisons
are therefore made between quintiles as defined in 2007–08 against quintiles as defined in 2011–12.
SOCIAL MARKET FOUNDATION
24
of care they use, they were paying 22% more for childcare in 2011–
12 compared to 2007–08, in real terms. And research by the Family
and Childcare Trust show that nursery costs for this age group have
continued to rise.11
Chart 3.4: Change in spending for children aged under two
70
60
50
40
30
20
10
0
2007–08 2011–12
Quintile 1 Quintile 2
£ pe
r wee
k pe
r chi
ld
Quintile 3 Quintile 4 Quintile 5
SMF analysis of Family Resources Survey 2007–08 and 2011–12.
Chart 3.5: Change in hours of childcare used for children aged under
two, 2007–08 to 2011–12
5.0
4.0
3.0
2.0
1.0
0.0
-1.0
-2.0
Free hours Paid-for hours
Quintile 1
Cha
nge
in h
ours
per
wee
k pe
r chi
ld
Quintile 2 Quintile 3 Quintile 4 Quintile 5
SMF analysis of Family Resources Survey 2007–08 and 2011–12.
11 Jill Rutter and Katherine Stocker, Childcare costs survey 2014 (London: Family and Childcare Trust, 2014)
RIDERS ON THE STORM
25
Non-essentials12 Households have reduced spending on leisure goods but not leisure
services. As shown in Chart 3.6, the drop in leisure goods spending
has been especially sharp for the middle and fourth quintiles. In the
case of leisure goods – which include products such as television
sets, phones, newspapers, books, games consoles and DVDs –
households have been helped by the fact that inflation has been
relatively muted. Prices for audio-visual equipment for instance
fell by more than 50% since 2007, general recreational items saw a
price rise of only 1.5% over the period from 2007 to 2011, and other
leisure goods saw a level of inflation in line with the general price
increase of about 14% during that period.13
In contrast, leisure services – which include sports clubs,
cinemas, live entertainment and leisure subscriptions and holidays
– generally saw inflation above or in line with the average, making
it harder to cut back, as shown in Chart 3.7.
Chart 3.6: Spending per week on leisure goods (2011 prices)
£40£35£30£25£20£15£10£5£0
Quintile 5
Quintile 4
Quintile 3
Quintile 2
Quintile 1
2011 2007
SMF analysis of Living Costs and Food Survey 2007 and 2011
12 Note for leisure costs, we rely on LCFS data. LCFS does not follow households longitudinally. Comparisons
are therefore made between quintiles as defined in 2007–08 against quintiles as defined in 2011–12.
13 ONS, “Consumer Price Inflation Reference Tables”, January 2014, http://www.ons.gov.uk/ons/rel/cpi/
consumer-price-indices/january-2014/consumer-price-inflation-reference-tables.xls.
SOCIAL MARKET FOUNDATION
26
Chart 3.7: Spending per week on leisure services (2011 prices)
£40 £60 £80 £100 £120 £140 £160 £180£20£0
Quintile 5
Quintile 4
Quintile 3
Quintile 2
Quintile 1
2011 2007
SMF analysis of Living Costs and Food Survey 2007 and 2011
THE FUTURE SQUEEZE
Overall, those in the middle today have been remarkably resilient in
changing their spending habits to suit their means. But in a sense
they are yet to be fully tested. Over two-thirds of middle quintile
households are homeowners, and just over half have a mortgage.
The high proportion of homeowners with a mortgage has helped
middle income households so far; but this also means they will face
rising costs when interest rates rise again.
A key determinant of the impact of this squeeze will be how
the Bank of England raises interest rates, and whether rates rise in
tandem with income growth. Bank of England analysis recently
showed that a rise in interest rates of 2.5 percentage points would
increase the proportion of households with high repayments
compared to incomes by 50% if incomes rise by 10%, and double
the proportion if incomes remained unchanged.14 Within the
middle income group, this squeeze is likely to affect 30–50 year
14 Philip Bunn, May Rostom, Silvia Domit, Nicoal Worrow, Laura Piscitelli, The financial position of British
households, (London: Bank of England, 2013)
RIDERS ON THE STORM
27
olds the most. This age group is more likely than older cohorts to
own a home with a mortgage rather than own outright.
Even during this period of low interest rates, over one in 10 of
middle income households in 2011–12 reported being behind on
rent or mortgage at least once in the past year, with households
with children under more pressure, as shown in Chart 3.8. Many
of these households are likely to have a significant amount left to
repay on their mortgages. Whilst data is not available on the middle
quintile specifically, across all households, the proportion with high
loan to value ratios was higher in 2013 compared to before the
financial crisis. Just under a fifth have a loan to value ratio of over
75% on their outstanding debt.15
Chart 3.8: Proportion of households behind on rent or mortgage,
2011–12
5% 10% 15% 20% 25%0%
Quintile 5
Quintile 4
Quintile 3
Quintile 2
Quintile 1
Households with children All households
SMF analysis of Understanding Society, 2011–12, Wave 3
Looking more widely, 8% of middle quintile households said
they defaulted on a loan, bill, mortgage or rent in the past month.16
Middle income households have also – in part – been relying on
15 Philip Bunn, May Rostom, Silvia Domit, Nicoal Worrow, Laura Piscitelli, The financial position of British households
16 Which? “Consumer Insight Tracker”, February 2014
SOCIAL MARKET FOUNDATION
28
credit to help fill gaps. 13% of middle quintile households took out
a loan or credit card, borrowed from friends or family, or used an
authorised overdraft in the last month.17
“ARE YOU BETTER OFF THAN YOU WERE FOUR YEARS AGO?”
Almost 60% of middle income households say they are “comfortably
off” or “doing alright”. Having weathered the downturn reasonably
well compared to other households, they are – on the face of it
– fairly comfortable. But they certainly do not feel better off than
they were back in 2007, as shown in Chart 3.9.
Chart 3.9: Proportion of households answering “How well are you
managing financially?”
0%
10%
20%
30%
40%
50%
60%
70%
80%
Living comfortably Doing alright
Quintile 3 in 2011–12 Quintile 4 in 2011–12
2011–12 2011–122007–08 2007–08
SMF analysis of British Household Panel Survey, 2007–08, Wave 17; Understanding Society, 2011–12, Wave 3
Quintiles are defined based on the 2011–12 income distribution. The chart compares responses of the same
group of households in 2007–08 and 2011–12.
They are less satisfied with the amount of leisure time they
have, and less satisfied with life overall. This pattern is hardly unique
to the middle, as shown in Charts 3.10 and 3.11. But interestingly,
our analysis shows that life satisfaction patterns do not change
17 Which? “Consumer Insight Tracker”, February 2014
RIDERS ON THE STORM
29
substantially depending on how middle income households’
relative position in the income distribution shifted, indicating
that feeling better off is not necessarily strongly driven by actual
changes in income. It may instead be more related to prospects
for the future: which in 2011–12 were likely to be looking gloomy
compared to 2007–08.
Chart 3.10: Life satisfaction (proportion of individuals satisfied minus
proportion of individuals dissatisfied)
70% 80%60%50%40%30%20%10%0%
Quintile 1in 2011–12
Quintile 2in 2011–12
Quintile 3in 2011–12
Quintile 4in 2011–12
2007–08
2011–12
2007–08
2011–12
2007–08
2011–12
2007–08
2011–12
2007–08
2011–12
Quintile 5in 2011–12
SMF analysis of British Household Panel Survey, 2007–08, Wave 17; Understanding Society, 2011–12, Wave 3
Chart 3.11: Satisfaction with amount of leisure time (proportion of
individuals satisfied minus proportion of individuals dissatisfied)
35% 40%30%25%20%15%10%5%0%
Quintile 1in 2011–12
Quintile 2in 2011–12
Quintile 3in 2011–12
Quintile 4in 2011–12
2007–08
2011–12
2007–08
2011–12
2007–08
2011–12
2007–08
2011–12
2007–08
2011–12
Quintile 5in 2011–12
SMF analysis of British Household Panel Survey, 2007–08, Wave 17; Understanding Society, 2011–12, Wave 3
SOCIAL MARKET FOUNDATION
30
4. HELPING MIDDLE INCOME HOUSEHOLDS
Middle income families do not feel better off even though – in
relative terms – they are generally the ones that have weathered
the downturn well. They have kept their jobs and cut spending by
switching to cheaper items and cutting back. But some of squeeze
may be still to come. Depending on the recovery of real wages,
many of these households could find it harder to manage their
finances in the years ahead.
Despite widespread focus on the “squeezed middle”, in reality,
these households are unlikely to see substantial additional state
help in the form of tax breaks and credits, given the competing
pressures on Government spending. But these households can be
helped to do more of what they have already done well: shopping
around, comparing prices and making their money go further.
In markets where there is significant choice and competition,
such as food, consumers have been able to manage price increases
by changing what they buy and switching. In other markets, this is
harder to do. In energy markets, despite bill increases, consumer
switching has actually been falling since 2009.18 Whilst consumers
have been saving money by using less energy, switching appears –
in comparison – to be an untapped source for cutting costs.
Our data shows substantial variation in the amount spent
per month on energy across households within the five quintiles,
as shown in Chart 4.1. Variation is in part due to differences in
consumption. But other statistics suggest that some households
are not getting the best deal possible. Data from DECC shows
that consumers could be paying anywhere between 12.79 pence
per kilowatt hour to 18.69 pence per kilowatt hour for electricity
depending on which tariff they pick and the payment method
18 Oliver Finlay, Illuminating the energy market (London: Social Market Foundation, 2013)
RIDERS ON THE STORM
31
they choose.19 Even within regions, bill sizes can vary substantially
depending on tariff.
Chart 4.1: Range in spending per week on energy across the
quintiles, 2011
£0
£10
£20
£30
£40
£50Quintile 1 Quintile 2
Quintile 4 Quintile 5
Quintile 3
£0
£10
£20
£30
£40
£50
SMF analysis of Living Costs and Food Survey 2011.
Excludes outside values, based on ONS fuel, light and power variable
Financial services is another area where consumers are
foregoing savings by not shopping around for the best deal.
Around two in five savers earn interest of just 0.5 per cent or less,
and one in five receive just 0.1 per cent or less a year.20
So what can be done? In some markets, even where consumer
engagement and switching is high, the market structure works to
constrain the benefits of choice and competition. Car insurance
provides a good example of a sector where consumers frequently
compare prices and switch. But on investigating, the Competition
Commission – now the Competition and Markets Authority – found
that the way in which claims of drivers not at fault were settled by
19 Department of Energy and Climate Change, Quarterly Energy Prices December 2013 (London: HMSO, 2013),
Table 2.2.3
20 Financial Conduct Authority, “Human Face of Regulation”, speech by Martin Wheatley, 2013, http://www.fca.
org.uk/news/speeches/human-face-of-regulation
SOCIAL MARKET FOUNDATION
32
insurance companies was driving up prices.21 Consumer action has
its limits if markets are not functioning properly.
In other areas, helping consumers to get better value means
helping them to do what they already do well in some markets,
where prices and quality are easy to compare. Whilst in some
markets, structural changes may be needed, evidence suggests
that there are more gains to be made from helping more consumers
to become more active.
COMPARING AND SWITCHING
In recent years, Government and regulators have moved to make
switching quicker, easier and more convenient across a range of
markets – energy, financial services and telecoms. This is a positive
step. But the gains from making the process of actually switching
smoother and faster are now close to exhausted. 85–95% of those
who do switch providers – across a whole range of markets – say
that the process was easy. It is only amongst those who have
not switched that there are still perceptions that switching is
difficult.22 Among the middle quintile specifically, the proportion
that perceive switching to be easy is higher than average – across
a range of products including banking and financial products,
energy, telephony and broadband.23
The big gains now are to be found in helping consumers easily
compare the level of savings they can make by switching. In energy
markets, for example, the strongest driver of consumer activity is the
level of savings that consumers expect to make. Even perceptions
of the expected length and difficulty of the switching process itself
21 Competition Commission, “CC seeks to reduce the cost of motor insurance”, 2013, http://www.competition-
commission.org.uk/media-centre/latest-news/2013/Dec/cc-seeks-to-reduce-the-cost-of-motor-insurance
22 Ofcom, The Consumer Experience of 2013 (London: Ofcom, 2013)
23 Which? “Consumer Insight Tracker”, January 2014
RIDERS ON THE STORM
33
have little effect in preventing consumers from moving providers.24
Government and regulators need to ensure that consumers have
access to the right information to make choices, and that it is easy
to use information that does exist to calculate likely savings from
switching.
GIVING PEOPLE ACCESS TO THE RIGHT INFORMATION
A significant area of spending for many middle income families is
on childcare. Websites exist that allow parents to search for nearby
childcare options in their area, and compare Ofsted inspection
reports where they are available. But comparing prices across
different providers is much harder, making it difficult for parents
to compare different types of childcare and decide which is best
for their needs. Adding pricing data to the information that already
exists on information such as Ofsted inspections could make the
process of choosing childcare easier for parents.
In the context of middle income households vulnerable to
increases in mortgage payments in the coming years, a key factor
will be how the Monetary Policy Committee decides to raise rates.
The Governor of the Bank of England has said when rates do rise,
they will do so gradually.25 The extent to which rates rise in tandem
with wage growth will be crucial.
But beyond this, there are other ways to create more choice
and ensure better value for money for consumers. There has been
increased interest from the Bank of England in mortgages that
would allow consumers to fix their rates for a longer period of
24 Miguel Flores and Catherine Waddams Price, “Consumer behaviour in the British retail electricity market”,
(University of East Anglia, 2013)
25 See, for example, Bank of England, “The economics of currency unions”, speech by Mark Carney, 2014,
http://www.bankofengland.co.uk/publications/Pages/speeches/2014/706.aspx
SOCIAL MARKET FOUNDATION
34
time, as is possible in other countries including the US.26 Another
way for households to manage the inevitable rate rises to come is
to compare offers across mortgage providers to get better deals
– whether on fixed or variable rates. There has been confusion
among consumers recently over whether providers offering tracker
mortgages are able to raise their rates in the absence of a Bank
of England rate rise. Lack of transparency and simplicity around
mortgage products will make it much harder for consumers to
manage their finances when rates rise again. It is vital that banks
and regulators ensure that there is sufficient clarity around these
products so that consumers can make informed choices. Measures
such as the implementation of the Financial Conduct Authority’s
Mortgage Market Review will help with this.
LIBERATING DATA
There is also much more to do on helping consumers easily manage
their spending, compare products and pick the best deals using
data that already exists.
Innovation in banking and use of financial data can play a
valuable role here. Mobile banking innovators have already started
to explore ways of using banking data to better help consumers
understand and analyse their outgoings using transactional
information on how they spend their money.27 The midata initiative,
which is designed to encourage companies to release consumer
data such as usage statistics and the development of apps that
make it easier to process this usage data and compare prices is also
very promising step.28 Building on midata, the proposal to include
26 Treasury Committee, “Bank of England Financial Stability report hearings”, 2013, http://www.parliament.uk/
business/committees/committees-a-z/commons-select/treasury-committee/inquiries1/parliament-2010/
bank-of-england-financial-stability-report-hearings/
27 Cormac Hollingsworth and Emran Mian, Branching out (London: Social Market Foundation, 2014)
28 Department for Business, Innovation and Skills, “Providing better information and protection for consumers”,
2014 https://www.gov.uk/government/policies/providing-better-information-and-protection-for-consumers/
supporting-pages/personal-data
RIDERS ON THE STORM
35
QR codes on energy bills could help dramatically increase the ease
and convenience of comparing providers using the customer’s
own usage data.29 Even larger benefits could come from services
that analyse data across the market to show individual consumers
how their spending compares to the average. For example, in
energy markets, knowing how much more they are paying per unit
consumed compared to other households in their local area could
make the potential savings much more visible.
In some cases, the market will develop these new services
and applications itself. But Government needs to remain aware of
the fact that in other cases, relying on voluntary participation by
companies will not be enough. The case of QR codes on energy bills
is a recent example of where relying on voluntary action by energy
companies has not worked, and Government is now consulting on
modifying energy company licences to have QR codes included as
part of energy bills.
There is likely to be a first-mover disadvantage from voluntarily
participating in schemes such as the provision of QR codes and the
midata scheme more generally. This is because the first company
to release usage data to its customers is likely to be subject to
more switching away to its competitors, creating a commercial
disadvantage. Given this dynamic, there is a strong rationale to
ensure that major players in the same industry act at the same
time, whether this is achieved through voluntary agreement or
compulsion.
Sectors where this scheme has the most beneficial applications
are, firstly, ones where consumers tend to have ongoing
relationships with suppliers, and therefore where it is highly likely
that data is already being collected; and secondly, where data on
29 Department of Energy and Climate Change, “QR codes on energy bills put consumers in control”, 2014,
https://www.gov.uk/government/news/qr-codes-on-energy-bills-put-consumers-in-control
SOCIAL MARKET FOUNDATION
36
consumption or usage is needed to compare offers across different
providers effectively. At the moment, Government is looking
to ensure that customers of current account providers, mobile
phone companies and energy providers benefit from greater
access to data on their usage. The Treasury recently announced
that the major current account providers will give customers their
account data in a simple, standardised format that can be used in
comparison sites.30
But there are other products and services that the midata
scheme could be expanded to. In telecoms, there are likely to
be benefits from expanding beyond mobile phone companies.
As mobile and landlines have increasingly become substitute
products,31 it would be beneficial for consumers to be able to
compare usage and tariffs across both the landline and mobile
calls they make. Broadband is another service where consumers – if
armed with better information on how much data they typically
use – could make a more informed decision as to which broadband
package and speed is best for them. Within the financial sector,
information on a consumer’s typical incomings and outgoings
could help consumers better understand what types of savings
products are right for them – for example instant access versus
notice accounts and regular savings accounts versus those where
the rate does not depend on how often money is deposited.
Outside the regulated sector, supermarkets, especially those with
loyalty cards, are likely to have a wealth of data that could better
help consumers manage grocery spending.
JUST PUSHING DATA IS NOT ENOUGH
Policymakers should also learn the lessons from price comparison
websites, which have made it easier to compare products, but
30 HM Treasury, Budget 2014 (London: HMSO, 2014)
31 Ofcom, International communications market report (London: Ofcom, 2013)
RIDERS ON THE STORM
37
do have limitations. Firstly, older consumers are less likely to use
these sites.32 Secondly, there have been concerns about how
well these sites capture data on actual pricing – as opposed to
offering comparisons based on headlines prices only, and whether
in some cases conflicts of interest may mean that sites do not
provide fully objective comparisons.33 And thirdly, the effect of
actively prompting consumers to compare prices should not be
underestimated: consumers say that renewal notices for services
such as car insurance act as a trigger for them to look at what
other providers are offering.34 Switching rates for car insurance are
36% – triple the rate for energy and seven times the rate for bank
accounts.
Innovative applications are likely to play a significant role in
making comparisons easier, and they may well also develop to act
as prompts for consumers, for example by keeping track of usage
and alerting consumers when deals that better fit their needs
come onto the market. But where this does not happen, other
measures need to be considered by Government, regulators and
industry, proceeding through voluntary agreement, or compulsion
if necessary. The model of renewal notices in car insurance offers a
guide to what could work well for other sectors. Receiving a letter
once a year as a prompt, especially if combined with an easy to use
system of comparing offers using the consumer’s own usage data
could be very powerful.
Given competing pressures on public spending, providing
substantial help to middle income households through tax
reductions or more cash transfers is difficult. As recent analysis has
shown, Government is still less than half way through its planned
32 Consumer Futures, Price comparison websites: consumer perceptions and experiences (London: Consumer
Futures, 2013)
33 Financial Conduct Authority, “The FCA launches review into price comparison websites”, 2013, http://www.
fca.org.uk/news/the-fca-launches-review-into-price-comparison-websites
34 Consumer Futures, Price comparison websites: consumer perceptions and experiences
SOCIAL MARKET FOUNDATION
38
spending cuts.35 But middle income households have shown
remarkable resilience so far, demonstrating that under the right
circumstances, and with the right information to hand they can
manage their finances and cope with being under pressure. Food
spending is one example, where they have shopped around to
keep down costs in spite of rising prices. By equipping consumers
to negotiate markets where it is currently harder to secure better
value for money, middle income households can be helped to
better manage their finances in the future. Making this happen will
require action from Government, regulators and industry.
35 Institute for Fiscal Studies, “Still not halfway there yet on planned spending cuts”, 2014, http://www.ifs.org.
uk/publications/7086
RIDERS ON THE STORM
39
ANNEX: METHODOLOGY
OVERVIEW
We use longitudinal data from the British Household Panel
Survey (BHPS) and Understanding Society (US)36 to follow specific
households over time. We look at how households’ position in
the income distribution has changed, and focus on where the
households currently in the middle and fourth quintiles have come
from.
Our definition of quintiles is based on original income before
taxes and benefits to understand underlying trends before
government transfers, however, we take into account gross income
including benefits when looking at the spending squeeze. We
look at income, employment, spending patterns and leisure as
indicators of how households have fared.
There are some areas where there is insufficient information
in the BHPS and US surveys. We therefore supplement the BHPS/
US analysis with additional cross-sectional analysis from the Family
Resources Survey (FRS) and the Living Costs and Food Survey (LCFS),
for data on childcare, leisure and energy. Because these datasets
are cross-sectional, any changes over time are based on quintiles as
defined in specific years, and do not follow households over time.
Our analysis is based on data from the latest waves of each of these
surveys, 2011–12 (BHPS/US and FRS), and 2011 (LCFS). We also use
the FRS to show a snapshot of how the income distribution has
changed between 2007–08 and 2011–12.
For inflation-adjusted comparisons, we use the Consumer Price
Index.
36 Understanding Society is an initiative by the Economic and Social Research Council, with scientific
leadership by the Institute for Social and Economic Research, University of Essex, and survey delivery by the
National Centre for Social Research.
SOCIAL MARKET FOUNDATION
40
ANALYSIS OF UNDERSTANDING SOCIETY AND THE BRITISH HOUSEHOLD PANEL STUDY
The main analysis presented in this report is based on data from
two longitudinal datasets, British Household Panel Survey (BHPS)
and Understanding Society (US). Beginning in 1991, the BHPS was a
UK representative survey that followed the same set of individuals
and households for eighteen consecutive years. The final wave of
BHPS, Wave 18, was completed in April 2009.
Launched in 2010 as the successor to the BHPS, US is a
UK representative longitudinal dataset that surveys the same
individuals and households over time. Understanding Society
incorporated almost 6,700 individuals that had previously been
in the BHPS sample. In total, the latest wave of Understanding
Society (Wave 3) contains data from 49,739 individuals and
27,783 households, and was collected between January 2011 and
December 2012.
The first stage of our analysis focused on this sample of
households from US. Of this sample, we excluded households that
had at least one member of pensionable age (defined as 60 for
women and 65 for men), as our analysis is focused on the working
age population only. Accordingly, approximately 33% of the original
household dataset was excluded and the resultant dataset was
composed of 34,656 individuals from 18,518 households.
When we conducted our analysis, derived data on original
income before taxes and benefits was unavailable. We therefore
constructed our own original income variable by subtracting
household monthly benefit income from the state from monthly
gross household income. We used this original income variable to
construct our income distribution and assign income quintiles to
the households in our dataset. We then analysed household and
individual characteristics within these groups to produce top-level
RIDERS ON THE STORM
41
statistics on what middle income households looked like in 2011–12
and how they differed from households in other quintiles.
For the second stage of our analysis, we tracked households
over time, comparing their situation in 2007–08 against that in
2011–12. Specifically, we matched households from Wave 17 of
the BHPS to households in Wave 3 of US. We excluded households
that split. We also excluded any households that contained an
individual of pension age in either one of the two waves to avoid
picking up any trends related to entering retirement. The resultant
sample consisted of 3,166 successfully matched households with
5,931 individuals.
As with the US dataset, we allocated households a quintile
based on the income distribution in the BHPS 2007–08 dataset.
This meant that each of our matched households had two quintiles
associated with them – one for 2007–08 and one for 2011–12.
In analysing characteristics across the two years, we made the
following decisions to allow comparability:
• Food: In contrast to US, BHPS denotes household expenditure
on food in ranges. We convert expenditure ranges to
actual monetary values by taking the midpoint of each
food expenditure band. To make food expenditure in BHPS
comparable to US, we derived variables in US and BHPS that
captured monthly expenditure on food and groceries as well
as food bought outside the home. This required summing
the two food variables in US, monthly household expenditure
on groceries and monthly household expenditure on
food bought outside the home (including restaurants and
takeaways) respectively. For BHPS, we aggregated monthly
expenditure on food consumed in the home and household
expenditure on food and groceries including takeaways.
• Energy: In both BHPS and US, household expenditure on
SOCIAL MARKET FOUNDATION
42
gas, electricity and other fuels are separate variables. We
generated a new variable that aggregates each component of
energy expenditure to obtain an estimate of household total
expenditure on energy in the previous 12 months.
• Mortgage and rent: In both US and BHPS, mortgage paid is
a variable in monthly terms whilst rent paid is a variable that
covers a range of time periods. Accordingly, we transformed
this rent variable to monthly terms using variables detailing
amount of rent paid in last instalment and weeks covered by
rent paid. We then generated a new variable that captures
both monthly mortgage and rent paid in last month.
ANALYSIS OF THE FAMILY RESOURCES SURVEY
The Family Resources Survey (FRS) is a cross-sectional dataset used
for the analysis of household incomes, inequality and household
expenditure. For the analysis in this report we used data for the
years 2007–08 and 2011–12 for childcare expenditure patterns and
for the analysis of general patterns in the change in the income
distribution between 2007–08 and 2011–12.
The total sample size of the survey is 20,759 households for
the year 2011–12 and 24,977 households for the year 2007–08. We
excluded households that had at least one member of pensionable
age (defined as 60 for women and 65 for men), as our analysis is
focused on the working age population only. This left a sample of
16300 households for 2007 and 13581 households for 2011.
We used the FRS to show how the UK’s income distribution
profile changed between 2007–08 and 2011–12. To do this, we
constructed household original income (before taxes and benefits)
to assign quintiles. We did this by aggregating gross income from
employment, self-employment, investment income, and other
non-benefit income.
RIDERS ON THE STORM
43
We also used the FRS for data on childcare. About 55% of
households did not contain any children (stable for both waves of
the FRS we looked at), so our analysis of childcare usage and childcare
expenditure was based on about 7,340 and 6,200 households in
2007–08 and 2011–12, respectively. We then computed the average
childcare cost by quintile and number of hours used, both of free
of charge childcare and of paid for childcare, per child for different
age groups. This calculation was based on summing expenditure
across different types of childcare for each child within the dataset,
as well as the number of hours used of both informal and informal
childcare.
ANALYSIS OF THE LIVING COSTS AND FOOD SURVEY
The Living Costs and Food Survey is a cross-sectional dataset
collecting information on the expenditures of households in
the UK. It has been collected since 2008 as a replacement to the
Expenditure and Food Survey.37 The survey had a sample size of
about 5700 households in 2011, and about 6100 in 2007. For this
report we have used data on expenditure on leisure goods and
services, and expenditures for internet and fuel.
As with the FRS, we excluded households with members of
pension age, leaving around 4,000 households in each year we
analysed. We grouped households into income quintiles using the
same methodology as for the FRS, first constructing an original
income variable including wages and earnings income, income
from self-employment, investment income and other sources, plus
income from private pensions.
Analysis of the expenditure on leisure goods and services and
sub-items within this group were then performed by computing
weighted averages across quintile groups.
37 For more information see UK Data Service, http://discover.ukdataservice.ac.uk/series/?sn=2000028.
Copyright © The Social Market Foundation, 2014
Copyright © Social Market Foundation, 2014
ISBN: 978-1-904899-89-1
£10.00
ISBN: 978-1-904899-89-1£10.00
SOCIAL MARKET FOUNDATION11 Tufton Street | Westminster | London SW1P 3QBPhone: 020 7222 7060 | Fax: 020 7222 0310 www.smf.co.uk
THE SO
CIA
L MA
RK
ET FOU
ND
ATIO
NR
IDER
S ON
THE STO
RM
: BR
ITAIN
’S MID
DLE IN
CO
ME H
OU
SEHO
LDS SIN
CE 2007
N
IDA
BR
OU
GH
TON
, ON
YIN
YE EZ
EYI, C
LAU
DIA
HU
PKA
U
In the years since the financial crisis, working age households have seen wage stagnation and rises in the cost of living, prompting widespread focus on the so-called “squeezed middle”. This report looks at the reality of middle income households coming out of the downturn – using the British Household Panel Study and Understanding Society to track specific households over time.
Middle income households have shown remarkable resilience. Many have taken on more work and increased their incomes. There is a surprising amount of movement across the income distribution, with many climbing up. Middle income households have also shopped around and cut back to reduce the effect of rising prices.
But for many of these households, the squeeze may be yet to come when mortgage rates rise again. And in some markets it is much harder to compare products and secure value for money. With competing pressures on public spending, these households can nevertheless be helped by supporting them to do what they already do well – shopping around, comparing prices and switching.